Getting Started

Planton is a free desktop app and CLI for your cloud infrastructure. You download it, open it, and deploy to your own cloud — with clean, auditable infrastructure-as-code running underneath. No account, no sign-up, no ceremony.

This page gets you from zero to a first deployment with the desktop app — the primary way to use Planton — and then shows how the CLI drives the same engine when you'd rather stay in the terminal.

Deploy with the desktop app

The fastest way to see Planton work.

1. Download Planton for your platform from planton.dev/download, and open it. On first launch, Planton starts a local instance on your machine — a local control plane that runs entirely on your own hardware. There is nothing to sign up for.
2. It finds the cloud you're already signed into. Planton detects your existing local credentials (for example, the AWS profile you already use) — no connection to configure.
3. Pick a stack and fill a short form. Choose a ready-made stack or a single component, fill in only what's truly required (smart defaults handle the rest), and review a plain summary plus the exact manifest that will be applied.
4. Click deploy, and watch it come online. See the architecture before you deploy, then watch each resource light up as it is created — real infrastructure-as-code, stored and versioned, every change a diff.

The desktop app deploys to AWS, GCP, Azure, and Kubernetes.

Or drive it from the CLI

The planton CLI is a companion to the desktop app — it drives the same engine and the same proven modules. The managed backend (state, ready-made charts, and history) comes from Planton itself, so most people run the app and reach for the CLI when they'd rather stay in the terminal.

It mirrors the two gestures Kubernetes made great — and frees them for every cloud. There are two commands, and they are never interchangeable:

• planton apply -f <manifest> applies a single component from one manifest — the kubectl apply -f parallel.
• planton chart install <chart> … installs a whole environment from a chart — the helm install parallel.

Install the CLI

go install github.com/plantonhq/planton@latest

Verify the installation:

planton version

Apply a single manifest

Every Planton manifest follows the Kubernetes Resource Model (KRM) — the same apiVersion, kind, metadata, spec shape you already know, extended to every cloud. Create a file named bucket.yaml:

apiVersion: aws.planton.dev/v1
kind: AwsS3Bucket
metadata:
  name: my-first-bucket
spec:
  awsRegion: us-east-1

Deploy it:

planton apply -f bucket.yaml

The CLI resolves the proven, pre-built module for AwsS3Bucket, runs it locally, and streams the live output to your terminal as the resource is created — genuinely live, not fire-and-forget.

Install a whole environment

A chart installs a whole environment in one command — many resources, wired together:

planton chart install aws-ecs --name api --env dev --values values.yaml

You'll watch every resource in the environment come up, live, as it happens.

What's running underneath

Whichever path you take, Planton runs proven, pre-built infrastructure-as-code modules for you — it does not ask you to write them. Each module is backed by Terraform or Pulumi, with secure, cost-aware, well-architected defaults baked in. Your configuration is stored and versioned, every change is a diff, and you can export it and run it yourself at any time. Nothing locks you in.

Next steps

• Understand the model. Read Desktop App and CLI for how the two commands map to Kubernetes, and Core Concepts for the resource model.
• Browse the catalog. Explore the components across 17 cloud providers, each with a guided form and ready-made presets.
• Go deeper. See Manifests, Dual IaC Engines, and State Management.
• Troubleshoot. Check the Troubleshooting Guide if you run into problems.